Secreted proteins direct the initiation, growth and patterning or the developing skeletal elements. BMPs are important examples of such factors. BMP family members were first identified over 15 years ago by their ability to induce bone in experimental settings. Yet surprisingly little has been learned about the precise roles different members of this family play during normal chondrogenesis and osteogenesis. This is a complex problem, as there is likely to be significant functional redundancy between different BMP family members. BMPs are likely to play multiple roles at distinct steps of skeletal formation making later stages harder to assess and moreover, loss of function of several of these BMP genes results in early embryonic lethality before skeletal phenotypes can be assessed.. We will take a genetic approach to this problem in mice, utilizing conditional alleles of BMP2 and BMP4 and a null allele of BMP7. Function of thse genes will be removed, singly and in combination, by expressing ere recombinase from promoters known to drive expression at various steps of skeletal formation. In particular, we will utilize an existing Prx1::cre line to remove BMP activity prior to chondrogenesis in the limb, and Coll:: ere to remove their activity in osteoblasts within the appendicular skeleton. Similarly, these gene activities will be removed in the neural crest-derived craniofacial skeletal progenitors with a Wntl::cre and Col::cre will again be used to study the roles of BMPs in dermal bone formation. The cellular events of dermal bone formation are very poorly understood. We will use the chick system to identify markers for different intermediate cell types during dermal ossification, determine when and where growth factors are expressed, and use viral misexpression in chick to study their roles in regulating dermal bone formation.